While Crohn's disease (CD) has been clearly identified as a Th1-mediated inflammation, the immunopathogenesis of its counterpart inflammatory bowel disease, ulcerative colitis (UC), has remained enigmatic. In previous studies we developed an experimental mouse model of inflammation known as oxazolone colitis that bears a strong histopathologic resemblance to ulcerative colitis. In further work we established that the inflammation in oxazolone colitis is associated with an initial IL-4 response that rapidly gives way to an IL-13 response and that the latter arises from NKT cells, since it is elicited by a-galactosylceramide, a glycolipid antigen that stimulates "invariant chain" NKT cells when presented to the latter in association with CD1d. Finally, we showed that the IL-13-producing NKT cells are the cause of oxazolone colitis, since deletion of CD1d cells by anti-CD1d antibody or blockade of IL-13 by IL-13Ralpha2-Fc prevents the colitis. Similarly, disease could not be induced in knock-out mice lacking CD1d or components of the invariant TCR recognizing a-galactocylceramide. Taken together, these data indicated that an experimental inflammation resembling ulcerative colitis could be caused by an IL-13-secreting NKT cell. In a logical extension of these studies we show in the present work that lamina propria (LP) T cells from UC patients produce significantly greater amounts of IL-13 than control cells and little or no IFN-g, whereas comparable cells from CD patients produce large amounts of IFN-gamma and only small amounts of IL-13. In addition, we identified T cells bearing an NK marker as the source of the IL-13, but showed that these cells were not NKT cells bearing an invariant TCR since they were not stained by a-galactocylceramide-loaded CD1d tetramers and could not be stimulated to produce IL-13 by a-galactocylceramide. Nevertherless, their designation as NKT cells (with non-invariant TCR?s) was assured by their ability to produce IL-13 following stimulation by a B cell transfected with high levels of CD1d. These studies indicate that UC is associated with an atypical Th2 response characterized by IL-13-producing non-invariant NKT cells. That this response is pathologic is strongly suggested by the prior studies of oxazolone colitis (described above) showing that NKT cells producing IL-13 is the cause of experimental colitis resembling UC. Overall, these studies clear the way to treatment of patients with ulcerative colitis with agents that block the IL-13 response.